The inheritance of traits from parent to offspring is a universal characteristic of life on earth, and has fundamental consequences for its inner workings, and for evolution. Recent results in the field of epigenetics have resurrected the once-discredited possibility that the environment of parents could have an effect on the phenotype of their offspring. Inheritance of acquired characters (ie passage of environmental information from one generation to the next) is often called Lamarckian inheritance, and demonstration of its existence would drastically alter how we think about evolution, and how human epidemiological studies are carried out. I have used a microarray approach to identify transgenerational effects of the paternal environment on offspring phenotype in mice, linking paternal low protein diet to cholesterol metabolism in offspring. In this project I propose to systematically characterize the mechanism by which environmentally-directed traits are inherited in mice. Based on preliminary data, it appears that low protein diet specifically alters the levels of specific small RNA fragments in sperm, and that after fertilization these RNA fragments can alter gene expression programs in the early embryo. Here, I propose a combination of genetic, epigenetic, cell biology, and developmental biology studies to investigate the mechanistic basis for generation of these RNAs in response to diet, and to determine the functional consequences of these RNAs for the next generation. These studies will have a revolutionary impact on fields ranging from reproductive biology to evolution to epidemiology.